Drug therapy for a wide range of common disorders, including epilepsy, depression, and cancer often produces adverse cognitive side effects that are disruptive to daily life. They can be particularly debilitating when an essential function, suc as verbal communication, is adversely affected. However, little is known about the mechanisms underlying a drug's impact on cognition and consequently why some individuals are more prone to experience drug-induced cognitive deficits than others. The long-term goal of this project is to enhance clinical strategies and inform drug development in order to maximize the benefits of individual medication therapy while minimizing adverse cognitive/language-related side effects. TPM, a second generation and broad spectrum antiepileptic drug, is an ideal, initial choice for our studies because, in addition to being associated with adverse effects on attention and memory, it has a unique cognitive signature affecting language use in a subset of patients. Moreover, the mechanism by which TPM affects cognition, including linguistic behavior, has not been well established. The objective of this application is to elucidate the relationship among TPM exposure as measured by plasma drug levels, its neurophysiological effects, and consequent effect on the cognitive processes observable in everyday language. In order to accomplish this objective, the tools of clinical pharmacology, computational linguistics, neuroscience, and engineering will be applied to the design and execution of randomized, double blind, crossover studies using several levels of TPM exposure in healthy adults. A novel system for automated language and speech analysis (SALSA) makes possible an objective and quantitative characterization of linguistic behavior as reflected in spontaneous speech. SALSA, EEG, and pharmacokinetic-pharmacodynamic analyses will be used to achieve the specific aims of this grant: 1. Characterize TPM-induced effects on linguistic behavior 2. Determine the neurophysiological effects of TPM on recognized ERP indices of verbal working memory load and 3. Predict individual vulnerability to TPM-induced impairments in linguistic, memory, and executive functions. Achieving these aims will offer insight into the mechanisms underlying drug-induced cognitive deficits. It will also lay the foundation for a new line of research that wil further delineate these mechanisms and provide methods to predict individual patient response. Having an understanding of the physiological mechanism(s) that underlie language use and their relationship to spontaneous speech will enable us to apply this methodology in conjunction with state-of-the- art brain imaging and pharmacokinetic-pharmacogenomic analyses to obtain a comprehensive picture of how cognitive-impairing and, later, cognitive-enhancing medications interact with individual characteristics. Future directions will extend this approach to patients across different disorders (e.g., epilepsy, Alzheimer's) and multiple drug classes (e.g., anticholinergics, antipsychotics).